© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 32861
    Prev
    Next
    INDEX
    Explanations

    self-defeating/destructive patterns or maladaptive variants

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    çŀ¬
    -0.11
    å´Ľ
    -0.10
     stro
    -0.09
    æ½®
    -0.09
     spont
    -0.09
    æī¶
    -0.09
    argas
    -0.08
     вн
    -0.08
    çľ¼åīį
    -0.08
    slack
    -0.08
    POSITIVE LOGITS
    å¾Īæ·±
    0.12
    æŧ
    0.10
    伤åı£
    0.10
    å·²ä¹ħçļĦ
    0.10
    家åºŃæķĻèĤ²
    0.10
    积累
    0.10
    ä»ĬçĶŁ
    0.09
    æįŁä¼¤
    0.09
    دة
    0.09
    è¿Ļè¾ĪåŃIJ
    0.09
    Activations Density 0.060%

    No Known Activations